When is more better? Modeling the effect of dose on the efficacy of the MAPT multidomain intervention as a function of individual characteristics. (31st December 2021)

• Record Type: Journal Article
• Authors: Belleville, Sylvie; Cloutier, Simon; Mellah, Samira; Willis, Sherry L; Vellas, Bruno; Coley, Nicola; Andrieu, Sandrine; Ngandu, Tiia
• Is Part Of: Alzheimer's & dementia. Volume 17(2021)Supplement 10
• Page Start: n/a
• Page End: n/a
• Publication Date: 2021-12-31
• Subjects: Alzheimer's disease -- Periodicals; Alzheimer Disease -- Periodicals; Dementia -- Periodicals; Démence; Maladie d'Alzheimer; Périodique électronique (Descripteur de forme); Ressource Internet (Descripteur de forme); 616.83
• DOI: 10.1002/alz.054948
• Languages: English
• ISSNs: 1552-5260

Abstract

Abstract: Background: Dose is a critical factor examined in pharmacological treatment. However, little is known about the optimal dosage for non‐pharmacological multidomain interventions as well as the function that relates dose to improvement. These gaps have been assessed in this present study in interaction with individual characteristics, which may moderate the effect of dose. Magnification models predict that healthier individuals will benefit more from prolonged practice, whereas reserve models predict that training will reduce cognitive disadvantage. Method: The data from the Multidomain Alzheimer Preventive Trial (MAPT) was used to determine the relationship between cognitive training dose and cognition, and its interaction with sex, age, education, entry criteria, CAIDE risk score, frailty and Apoe4 status. The MAPT is a three‐year randomized controlled trial, which includes cognitive training, physical activity, nutrition and omega‐3 polyunsaturated fatty acids in at‐risk individuals. The number of hours of cognitive training attended (max.=28) was used as a measure of dose in the 749 participants randomized to the multidomain intervention. Result: First, polynomial regression analysis modelled cognitive performance over time on memory, verbal fluency and composite cognitive score. Mixed model analyses measured the effect of dose on expected scores, and interaction with individual factors were assessed. A cubic function explained the relationship between dose and performance indicating benefits up to 12‐14 training hours followed by a plateau. Interactions were found with individual characteristics. Older participants and those with a higher CAIDE score reached their plateau more rapidly. Individuals with lower education and frail individuals benefited more from increased exposure. Conclusion: A non‐linear dose‐response function indicates that more is not necessarily better. Here, the optimal dose was found to be 12‐14 hours of training, about half of what was potentially available to participants. A magnification effect is found for age and vascular risk factors as younger and healthier participants show larger benefit. However, a reserve effect is found in frail or less educated participants as they benefitted more from additional training experience than their more privileged counterparts.